British Institute of Persian Studies
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY, SOUTH IRÁN
Author(s): Akira Tsuneki, Mohsen Zeidi and Katsuhiko Ohnuma
Reviewed work(s):
Source: Irán, Vol. 45 (2007), pp. 1-22
Published by: British Institute of Persian Studies
Stable URL: http://www.jstor.org/stable/25651409
Accessed: 24/07/2012 01:11
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http ://www.j stor.org/page/info/about/policies/terms .j sp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
British Institute of Persian Studies is collaborating with JSTOR to digitize, preserve and extend access to Irán.
STOR
®
http ://www.j stor.org
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY, SOUTH IRÁN
By Akira Tsuneki, Mohsen Zeidi and Katsuhiko Ohnuma
University of Tsukuba, Iranian Center for Archaeological Research, Kokushikan University
Abstract
The Eastern Zagros region has not previously been considered in the study of Neolithisation in the Near East. Our
recent research, which was carried out as one of the salvage projects in the Bolaghi valley, Fars, sheds new light in
this area. Two cave sites were excavated and both sites produced a considerable amount of lithic artefacts. They show
a coherent assemblage which appears to dáte to some point within the "Proto-Neolithic" framework of the Zagros
region. Though the subsistence remains háve not yet been fully studied, our evidence may bridge the hiatus between
the end of the Epi-Paleolithic and the beginning of the Portery Neolithic in the Eastern Zagros.
Keywords
Proto-Neolithic; Eastern Zagros; the Bolaghi valley; lithic artefacts; Neolithisation.
I. INTRODUCTION
The study of the transition from the Epi-Palaeolithic to
the Neolithic is very significant in human history
because it deals with the shift from food gathering to
food producing societies. Most of the archaeological
research on this transition has been carried out along the
hillsides of the Fertile Crescent in the Levant, southern
Anatolia, northern Mesopotamia, and the western
Zagros Mountains, and areas peripheral to this háve
received little attention. Although the natural
environment is not so different from that of the western
Zagros, the eastern Zagros Mountains is one of these
latter areas. Many Epi-Palaeolithic caves and shelters
and Pottery Neolithic settlements háve been discovered
and registered in Fars province, especially in the
Marvdasht plain (e.g. Rosenberg 1985; 2003; Alizadeh
2006). However, sites dating between the Epi-
Palaeolithic and Pottery Neolithic periods háve not
previously been reported. The fact that this transition era
is still unknown in Fars province suggests that the
eastern Zagros did not play a role in this transition.
However, our recent investigations háve revealed the
existence of occupation that may dáte to this transition-
al phase in this region. We tentatively call this transition-
al period the "Proto-Neolithic". A suitable defmition for
this period in this region has not previously been
proposed and here, this term is ušed to indicate a
chronological phase and does not suggest the existence
of domestication.
Our investigation was carried out as one of the
salvage projects in the Sivand dam area (ICAR 2006).
The new Sivand dam has been constructed at the
southern end of Darr-ye Bolaghi, which is one of the
small basins in the eastern Zagros Mountains. The dam
site is about 30 km. north-east of Persepolis and about
12 km. south-west of Pasargadae (Fig. 1). A narrow
valley, named Tang-i Bolaghi, and most of the other
parts of this small basin will be flooded by the construc-
tion of this dam. Hereafter, we refer to both the valley
and basin together as the Bolaghi valley. We carried out
the first seasoďs investigations from mid-July to the
beginning of August 2005. Although we conducted
some surveys during our first season of investigations,
we report here only the operation of the excavations at
two caves that produced relevant materials.
The sites within the dam area were initially surveyed
and numbered by Mr 'Atai and ICAR from BV1 to
BV129 (' Atai 2003; as directed by the Iranian Center for
Archaeological Research the site codes háve now been
changed from BV to TB = Tang-i Bolaghi). Most of the
sites were cairns, open-air sites, and sites with evidence
of metallurgic activity, and dáte to the Achaemenian,
Sasanian and early Islamic periods. Although some
sites dáte back to the Chalcolithic period, very few
prehistorie open-air sites were discovered. We concen-
trated our attention on the caves and shelters along the
skirt of the surrounding mountains. Within the dam area
many caves and shelters were recognised. Based on the
results of our initial short visit, we chose two cave sites,
JOURNAL OF PERSIAN STUDIES
KUHE BOLAGHI
TANG-I BOLAGHI
Pasargfrdaé^
Sa adatabad
,
KUHE CENAF
KUHE SIAH
Persepolis A.
Táli Bakun A
KUHE RAHMAT
^
Fig. 1. Tang-i Bolaghi and its neighbouring area.
N
fc
A caven and aheller*
# openair sítes
■ graves
x meteílurgic spots
10km
Fíg. 2. Sivand dam salvage area, showing important archaeological sites (afíer 'Atai 2003).
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
Fig. 3. Distant view ofTB75.
Fig. 4. Outlook from TB75.
TB75 and TB130, for the fírst seasoiťs excavations.
Although TB75 was already known, TB130 is a newly
discovered site that was identifíed by us. Both TB75 and
TB130 are located on the southern slope of Kuh-e
Bolaghi Bozorg, along the northern fringe of the
Bolaghi Valley (Fig. 2). The distance between these sites
is less than 2 km. The excavations unexpectedly
produced fruitful results, and shed new light on the
prehistory in the eastern Zagros region.
II. EXCAVATIONS AT SITE TB75
(ESHKAFT-E HÁJI BAHRAMI)
Site TB75 is a limestone cave that opens to the South-
west. The cave is known locally as Eshkaft-e Háji
Bahrami. It is the most conspicuous cave in the Bolaghi
valley, and comes into view immediately as you enter the
valley fřom the south (Fig. 3). From the cave location,
the centrál part of the Bolaghi valley is visible with a
JOURNAL OF PERSIAN STUDIKS
Fig. 5. Morphological
25m
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
partial view of the south (Fig. 4). The altitude of the
entrance to the cave is 1875 m. (Fig. 5). The opening
measures c. 9 m. wide and 2.8 m. high, and the depth to
the back of the main chamber is 19 m. Beyond this, the
cave continues and climbs to the west, more than 15 m.
(Fig. 6). The front terrace runs down to the small wadi,
named Tang-e Jilli. The height difference between the
cave and the wadi is over 30 m., and the terrace slope is
relatively steep, having an angle of 20°. The large
limestone rocks that háve fallen down from the rear
limestone mountain are distributed on the terrace slope.
Islamic period potsherds and flint artefacts are scattered
Fig. 6. CaveofTB75,
JOURNAL OF PERSIAN STUDIES
within the cave and on the terrace slope. In order to
clarify the cultural sequence both within and outside of
the cave, a small trench, measuring 2 x 1 m., was dug in
each location (Fig. 5).
II. 1. Trench A
A 2 x 1 m. trench was dug inside the cave, close to the
front. The trench was located c. 2 m. inside from the line
of the opening eaves and c.1.5 m. from the western wall
of the cave (Fig. 6). The trench was excavated to a
maximum depth of 1.5 m., but we did not reach the
bedrock at this depth. Though the cultural deposits were
divided into nine layers, they could be summarised into
the following three cultural phases. The uppermost
phase, consists of a series of ashy and brown soil layers
0.3 m. thick, and produced some Islamic glazed ware
sherds, that most likely belong to the Islamic period. A
hearth surrounded by cobbles was the only solid feature
in this phase. The middle phase consists of many burned
and grey soil layers, of about 0.2 m. in thickness. Some
deep pits were dug from the level of this phase into the
underlying deposits. These pits produced a great amount
of materiál that dates from the Achaemenid era. The
lowest phase, lower than the 0.5 m. from the surface of
the cave floor, produced only prehistorie materials,
especially lithic artefacts that were very similar to those
found from Trench B. The characteristics of these lithic
artefacts show that they belong to the period between
the end of the Epi-Palaeolithic and the beginning of the
Pottery Neolithic, i.e. the "Proto-Neolithic" as defmed
here. No potsherds were recovered in these lowest
levels.
112. Trench B
Another 2 x 1 m. trench, sunk in the middle of the
terrace slope, was dug 36 m. from, and along, the north-
south axis of Trench A, between the 1863 m.-l 862.5 m.
contour line (Fig. 5). We reached the bedrock at 1.35 m.
below the slope surface on the northern side of the
trench and at 0.95 m. on the southern side. Therefore,
1863.00m
Surface layer
1862.50m
1862.00m
1863.00m
1862. 50m
1862.00m
Fig. 7. Eastern section of Trench B at TB75.
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
the cultural deposits are about 1 m. thick in this part of
the terrace slope. The deposits could be divided into
three layers based on the colour and conditions of soil
(Fig. 7). The uppermost, Layer 1, is ashy brown,
measuring about 0.3-0.4 m. thick and containing
whitish ash. The next layer (Layer 2) is 0.2-0.3 m. thick
and is a reddish brown deposit, containing limestone
pebbles. A crude floor-like feature made of limestone
pebbles was observed at the bottom of this layer.
Although the limestone pebbles are quite sparse for a
stone floor, the terrace was partly levelled at the floor
level, indicating that this structure seems to háve been
cultural. The lowest layer (Layer 3) is a red-brown soil
with many limestone pebbles. It is the thickest deposit,
about 0.4-0.5 m. thick, accumulated on the bedrock soil.
Large pieces of limestone were spread unevenly
throughout the deposit. Except the crude floor in Layer
2, no other structural remains were discovered.
II. 3. Cultural Finds
In Trench A, the uppermost phase produced a limited
amount of Islamic glazed ware potsherds. The second
phase produced a great amount of large Achaemenid
storage jar fragments. These jars háve an orange-red
fabric, and are grit and grog tempered, well fired, and
sometimes háve a grey slip on the exterior surface.
Some body sherds háve raised decoration. The lowest
phase did not produce any potsherds. Fourteen
potsherds were recovered from Trench B, and all of
them came from the surface and Layer 1. Although
some of them were prehistorie potsherds, most
specimens were historie, especially Achaemenian. The
number of potsherds is quite small, particularly
compared with lithic artefacts recovered. It seems
likely that Layer 1 contains materiál that has been re-
deposited from higher up the slope. However, Layers 2
and 3 contain only "Proto-Neolithic" artefacts. It means
that these layers were the originál deposits or if re-
deposition oceurred, it oceurred during the Proto-
Neolithic period.
Trench A and B produced many lithic artefacts,
whose characteristics are quite similar. The former
produced 270 pieces and the latter produced 1583
pieces. Various kinds of green, grey, black and red
chert-like flints were ušed as raw materials. A small
number of obsidian microblades were also discovered,
however no obsidian cores or débitage were found. The
total amount of lithic artefacts unearthed from TB75 is
summarised in Table 1. The most numerous chipped
stones are microblades, measuring 2-5 mm. wide and
less than 30 mm. long (Fig. 8: 1-3). Most of them háve
minuté retouch or nibbling on one or both sides. This
evidence indicates that these tiny microblades were
ušed as elements of composite tools. Dozens of
microblade cores were discovered from both trenches.
They show various stages, from the initial pebbles to
the finál discarded cores. The most prevalent core types
are single platform conical and prismatic shaped cores
(Fig. 8: 5), regularly detached by pressure-flaking. In
addition to cores, microblade core rejuvenation flakes,
such as core tablets and bottom flakes, were frequently
recovered. The presence of so many cores makes it
likely that a large number of microblades were manu-
facrured at TB75. The tools made from blades include
backed pieces (Fig. 8: 4), notches (Fig. 8: 6), burins
(Fig. 8: 7) and geometrie microliths. For the production
of larger and heavier tools, such as end-serapers (Fig. 8:
8) and thumbnail-scrapers (Fig. 8: 9), both flakes and
TABLE 1. Total Amount of the lithic artefacts unearthed from TB75 and TB130.
TB75
Trench A
Trench B
TB130
Trench A
Trench B
Trench C
Trench D
Trench E
1.853 pieces
Surface (11), Layer 1 (4), Layer 2 (4), Layer 3 (0), Layer 4 (2), Layer 5 (3),
Layer 6 (0), Layer 7 (141), Layer 8 (47), Layer 9 (58)
Surface (3), Layer 1 (31), Layer 2 (404), Layer 3 (1,145)
3.419 pieces
(2)
Surface (83), Layer 1 (14), Layer 2 (41), Layer 3 (73), Layer 4 (403)
Surface (2), Layer 1 (6), Layer 2 (116), Layer 3 (215), Layer 4 (238)?
Surface (25), Layer 1 (145), Layer 2 (524), Layer 3 (492)
Surface (90), Layer 1 (473), Layer 2 (410), Layer 3 (67)
JOURNAL OF PERSIAN STUDIES
9
1.1 fi
P
i ' r
5 cm
^>
Fig. 5. Zií/iic artifacts from TB75 (1-5, 8: Trenek A—lowest phase, 6-7, 9: Trenek B—layer 3, 10-11: Trenek B—layer 2).
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
blades were ušed. Pointed pieces (Fig. 8: 10-11) were
also discovered, although the number is quite limited.
The inventory of lithic artefacts from Layers 2 and 3 of
Trench B is shown in Table 2. It is immediately evident
that microblades and the by-products of microblade
manufacture comprise the majority of the chipped stone
industry. The characteristics and chronology of this
industry will be discussed below.
III. EXCAVATIONS AT SITE TB130
Site TB130 is located about 1.5 km. south-east of
TB75. This shelter opens to the south-east and the view
from the cave is much less open than TB75. The
altitude of TB130 is 1848 m. and the height from the
wadi bed is about 30 m. (Figs 9 and 10). The depth of
TB130 is not so great. The shelter is 9 m. wide by 6 m.
deep, and the height of the opening is about 9 m. (Fig.
12). The front terrace is steep like that of TB75, but
relatively thick unnatural deposits were observed on the
slope. Many small chipped stones were scattered on the
terrace surface between the shelter and the wadi, and
they indicated the existence of Epi-
Palaeolithic/Neolithic cultural layers. One 2 x 2 m.
trench (A) was sunk inside the shelter, and two 2 x 1 m.
trenches (B, C) were sunk in the middle of the terrace
slope along the north-south axis. Another 2 x 1 m.
trench (D) was sunk 4 m. east of Trench B. As we
discovered part of a stone floor in this trench, another 2
x 1 m. trench (E) was dug just north of Trench D.
Therefore, the total excavated area measures 12 m. 2 . All
of these trenches, except Trench A, produced a consid-
erable number of chipped stones.
IV. TRENCH A
A 2 x 2 m. trench was dug near the back of the shelter.
However, we reached bedrock at a depth of only 0.3 m.
from the shelter floor. The trench was full of white grey
soft ash, modern animal dung and limestone pebbles.
We did not find any archaeological objects in this
trench, except a few modern potsherds. Therefore, we
can conclude that either the shelter was only ušed by
modern sheep herders, or that the old cultural deposits
háve been completely eroded away from the shelter.
IV 1. Trench B
A 2 x 1 m. trench was sunk into the middle of terrace
slope at 1832.8-1832.2 m. altitude. The thickness of
cultural deposits was about 0.7-0.6 m., and we reached
the reddish terra-rossa brown virgin soil at an altitude of
1832.2-1831.5 m. The deposits consist of four layers
(Fig. 13). The uppermost layer below the surface is a
grey-brown layer, followed by a whitish limestone
pebble layer, a grey ashy layer, and finally a reddish
grey layer which is above the natural deposits. All of the
layers were accumulated on top of the surface slope, and
we could not recognise any structures or activities that
Fig. 9. View oj TB 130.
10
JOURNAL OF PERSIAN STUDIES
TABLE 2. Inventory of lithic artefacts from Layers 2 and 3 of Trench B, TB75. (Microblades in the present
analysis are defíned as blades with the width less than 10 mm. Fragments are broken flakes or the flakes with no
clear feature. Core fragments are broken cores.)
Lithic Inventory
TB75 TB75
Trench B Trench B
Layer 2 Layer 3
TB130 TB130
Trench D Trench D
Layer 2 Layer 3
debitage pieces
cortical flakes
partially-cortical flakes
non-cortical flakes
flakes
crested flakes
microblades
5
4
12
1
4
2
12
13
13
broken debitage pieces
flakes
blades
crested flakes
microblades
234
1
1
126
1
104
core rejuvenation flakes
core tablets
core tablets for microblade cores
change of orientation flakes
broken core bottom flakes
modification flakes of flaking surface
12
1
1
2
5
chips or retouch-flakes
36
406
burin spalis
edge-rejuvenation flakes of end scrapers
303
2
290
retouched pieces
retouched flakes
5
2
1
retouched small flakes
2
retouched blades
1
2
retouched core tablets
1
notched pieces
1
notched blades
1
notched pointed blades
1
backed blades
2
1
pointed small flakes
4
1
side-scrapers
2
2
4
1
transversal scrapers
1
1
end-scrapers
9
4
2
steep scrapers
3
1
1
thumb-nail scrapers
1
5
5
2
burins
1
1
ušed obsidian pressure micro-blades
1
retouched small flakes
4
5
retouched blades
2
2
notched small flakes
1
side-scrapers
1
end-scrapers
3
thumb-nail scrapers
1
broken retouched pieces
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
11
microlithic tools
retouched microblades
notched microblades
lunates
atypical lunates
end-scrapers
burins
broken microlithic tools
retouched microblades
inversely-retouched microblades
notched microblades
percussion-flaked cores
pressure-flaked cores
irregular cores for 1 3
conical cores for flakes
prismatic cores for blades
conical cores for m 4
semi-conical cores for microbla
cylindrical cores for microblade:
3
5
1
1
2
2
2
3
1
1
pyramidal cores for microblades
prismatic cores for microblades
1
1
1
fragments
238
244
27
68
core fragments
61
128
21
15
total
404
1,145
524
492
levelled the slope. These layers, especially the lowest
one, produced a large amount of lithic artefacts.
IV. 2. Trench C
A 2 x 1 m. trench sunk into the terrace 2 m. south of
Trench B, at an altitude of 1831-1831.5 m. The
thickness, inclination, and characteristics of the cultural
deposits are quite similar to those of Trench B.
Moreover, the characteristics and composition of the
lithic artefacts are very similar.
trench, we could not determine the shape of the feature.
However, the northern edge of the floor seemed to be
fringed with a row of large limestones. The floor surface
is very rough, and it was probably ušed for working
purposes, such as butchering or cooking, rather than for
habitation. This stone floor was discovered in the third
layer beneath the slope surface. All of these layers
produced similar kinds of chipped stones. The thickness
and characteristics of these three layers are also similar
to those of Trenches B and C. As we stopped digging at
the stone floor level, we did not reach virgin soil in
Trenches D-E.
IV 3. Trenches D and E
Trench D was sunk into the terrace 4 m. east of Trench
B. Therefore, the altitude of this trench is almost the
samé as that of Trench B. As we encountered part of a
stone floor in this trench, another 2 x 1 m. trench, Trench
E, was dug just north of Trench D. These two trenches
formed one 2 x 2 m. trench. About 0.5 m. below the
surface, a rough stone floor pavěd with many angular
limestone pebbles was revealed (Figs 11 and 14). The
section shows that this stone floor was laid down on a
slightly levelled slope. As the floor extended out of the
IVA. Cultural Finds
As mentioned above, all of these trenches, except
Trench A, produced a considerable number of lithic
artefacts (N = 3419; see Table 1). The inventory of lithic
artefacts from layers 2 and 3 of Trench D is shown in
Table 2. The differences between the trenches and layers
are quite small, and it can be said that the lithic artefacts
of every layer belong to the samé industry.
Microblades (Fig. 15: 1-5) are the most numerous
artefact category among the chipped stone industry. As
for retouched tools, small scrapers, including side-
12
JOURNAL OF PERSIAN STUDIES
Fig. 10. Morphological map ofTB130.
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
13
Fig. 11. Stone floor discovered in Trenches D-E at TB130.
Fig. 12.ShelterofTB130.
14
JOURNAL OF PERSIAN STUDIES
1833.00m
1833.00m
1832.50m
Surface layer
1832.00m
1832.50m
1832.00m
Fig. 13. Eastern section of Trenek B at TB130.
serapers, end-serapers, steep-serapers and thumbnail-
serapers (Fig. 15: 6-11) are the most common. Backed
blades (Fig. 15: 12), geometrie microliths including
lunates and trapezoids (Fig. 15: 13-15), notches (Fig. 15:
16) and burins (Fig. 15: 17) were also recognised among
the tools. Single platform conical, pyramidal and
prismatic microblade cores were discovered in consider-
able numbers (Fig. 16: 1-6), and core rejuvenation flakes
such as core tables (Fig. 16: 7-8) were also numerous.
Besides the lithic artefacts, a few potsherds were also
recovered. Trench B produced 14 sherd fřagments. Four
of them are probably Neolithic potsherds, including one
typical Mushki painted ware rim sherd with light brown
fabric, straw temper, painted in black on red slip. The
other three Neolithic sherds are coarse plain ware. The
other sherds are much later specimens. Trench C
produced 22 sherds. Some Lapui red sherds were recog-
nisable among them. These are rim and body sherds with
buff-orange fabric, sand-tempered, red-slipped, and
burnished on both surfaces. Most of the other potsherds
are historie and modern specimens. Trenches D-E did
not produce any potsherds. Almost all of the potsherds
discovered in Trenches B and C came fřom the surface
and upper two layers. Compared with the number of
lithic artefacts, the number of potsherds recovered is
extremely small. Moreover, on the basis of the relative
dating of all of the materiál, this portery, even the Mushki
specimen, must dáte to a period later than all of the lithic
artefacts. As with the materiál from TB75 Trench B, it
seems likely that the upper layers at TB130 Trenches B
and C that contain a mixture of portery and lithic materiál
háve been re-deposited from higher up the slope. It is
possible that the lithics from the layers without portery
could also háve been re-deposited. However, it is notable
that there is consistency in the lithic assemblages from
the aceramic levels in each trench. This suggests that the
lithic materiál from these lower layers may represent a
coherent assemblage.
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
15
1832.50m
1832.00m
Surface layer
1832.50m
1832.00m
Fig. 14. Stone floor and western section ofTrenches D-E at TB130.
16
JOURNAL OF PERSIAN STUDIES
D
5 cm
ZE
Fig. 15. Lithic artifacts from TB130 (1-5: Trenek B — layer 4, 6: Trenek D — layer 2, 7-8: Trenek C — layer 2, 9: Trenek
B — layer 3, 10-11: Trenek B — layer 4, 12: Trenek B — layer 2, 13: Trenek C — layer 4, 14-15: Trenek E — layer 2, 16:
Trenek C — layer 3, 1 7: Trenek D — layer 2).
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
17
5 cm
Fig. 16. Lithic artifacts from TB130 (1, 3: Trenek D — layer 3, 2: Trenek D — layer 2, 4-5: Trenek B — layer 4, 6:
Trenek D — layer 3, 7: Trenek D — layer 2, 8: Trenek B — layer 4).
JOURNAL OF PERSIAN STUDIES
V. CHARACTERISTICS AND CHRONOLOGICAL
POSITION OF THE LITHIC ARTEFACTS
FROM SITES TB75 AND TB130
In this section we summarise the techno-typology of
the lithic artefacts unearthed at TB75 and TB130 in the
2005 fíeld season. A hypothetical placement of these
lithic artefacts into the lithic industries of the
surrounding regions is also proposed. The lithic
artefacts from TB75 amount to 1,853 specimens and
include 769 fragmentary pieces, and those from TB130
amount to 3,419 specimens including 896 fragmentary
pieces (Table 1). The raw materiál for these lithic
artefacts is chert-like flint, varying in colour from dark-
brown to green. There are no examples of sheen-
bearing pieces. Besides the flints, very few obsidian
microblades are visible among the lithic artefacts. The
characteristics and all attributes of these lithic artefacts
from both sites are similar to each other. Based on the
typology and technology of the artefacts found in the
earliest layers at each site, it is likely that both
assemblages belong to the samé chipped stone industry
and dáte roughly to the samé period. Therefore, we will
discuss the characteristics of both lithic assemblages
together. The largest of all the specimens unearthed is
an "initial" core, measuring 47 mm. in length, 24 mm.
in width and 24 mm. in thickness. The size of this
largest specimen strongly suggests that the manufacture
of lithic artefacts at the sites was initiated based on the
use of small sized raw materiál.
Complete specimens (unfragmented) total 3,607.
They are classified into débitage, core rejuvenation
flakes, chips or retouched flakes, retouched pieces and
cores (Table 2). The débitage, defmed as intentional
flake products of core reduction, are classified into
cortical flakes, partially-cortical flakes and non-cortical
flakes, the last of which are further classified into flakes,
blades and microblades. The core rejuvenation flakes
consist of core tablets (Fig. 16: 7-8), core bottom flakes,
flakes removed from the flaking surface of cores, and
change of orientation flakes. The chips are defmed as
byproduct flakes, and the retouch-flakes are derived
from retouch modifícation. It is rather difficult to differ-
entiate between these two categories unless they are
refitted. Included in the retouch-flakes are burin spalls
and edge-rejuvenation flakes of end-scrapers. The
retouched pieces consist of tools and/or weapons such as
backed pieces (Fig. 15: 12), notched pieces (Fig. 8: 6,
Fig. 15: 16), side-scrapers, end-scrapers (Fig. 8: 8),
steep-scrapers (Fig. 15: 7-11), thumbnail-scrapers (Fig.
8: 9) and burins (Figs 8: 7, 15: 17). Most of these
retouched pieces are very small, and many of them can
be classified as microliths, that is, backed, retouched and
notched microblades. Added to the microblades are
geometrie microliths of the lunate (Fig. 15: 13-14) and
trapezoid (Fig. 15: 15) categories, though small in
quantity. Because not a single micro-burin was
unearthed, it is almost certain that the geometrie
microliths were manufactured at both sites by
retouching on microblades without using the micro-
burin technique. This suggestion is supported by the size
of the microblades and the geometrie microliths
themselves. The absence of the micro-burin technique
may either indicate its disappearance in the region over
time or simply implies that this technique did not
originally exist there.
Many of the microblades, 2 to 5 mm. in width, bear
very fine, minuté retouch, probably having resulted
from use (Figs 8: 1, 15: 3,5). It seems, therefore, that
they were detached from cores and ušed without retouch
modifícation. Most of the microblades exhibit
extremely regular shapes, and the manner to detach
these microblades from the core was defmitely pressure.
There are very few traces of cresting, and it seems that
in most cases microblades were detached following
ridges on cores left by the removal of cortical flakes at
the very beginning of core reduction, instead of
following crest ridges prepared prior to the initial
detachment of microblades. The shapes of the cores are
conical, pyramidal and prismatic (Figs. 8: 5, 16: 1-6),
but interestingly enough not a single example of a
"bullet" core was unearthed. It is worthy to notě that no
pressure-flaked blades, very often seen among the later
Neolithic industries in the region, were unearthed. Also
noteworthy is the absence of heat treatment, a techno-
logical choice which seems to háve been adopted in the
region during the Neolithic period, to improve the
quality of raw materiál for making tools or weapons by
pressure.
In overall features, the lithic materials from the sites
TB75 and TB130 composed of tool types, such as end-
scrapers, thumbnail-scrapers, non-geometric microliths
(retouched, backed, notched microblades) and
geometrie microliths seem to retain the characteristics of
the Epi-Palaeolithic Zarzian in the Zagros Mountains,
best represented at the rock-shelter site of Warwasi in
West Irán (Olszewski 1993). The Zarzian assemblages
at Warwasi were analysed by Olszewski, who grouped
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
19
them into four stratigraphical units. According to
Olszewski, the lithic artefacts commonly seen in these
units are non-geometric microliths, notch/denticulates,
and thumbnail-scrapers. Micro-burins appear in Unit 2
(the second earliest unit) in association with geometrie
microlithis (Olszewski 1993). Illustrations of blades and
microblades demonstrate that pressure flaking was not
employed in their detachment from single-platform or
opposed-platform cores (Olszewski 1993: 208; Fig. 8:
1). Contrary to the Zarzian assemblages at Warwasi,
elaboráte pressure flaking was employed at TB75 and
TB130 to detach microblades. This is a major techno-
logical element that differentiates our materials from the
Zarzian industries.
Although differences in the artefact inventories
between TB75 and TB130 are now under study as are
inter-layer differences, both materials might be
tentatively placed within the temporal framework
spanning from the microlith dominated "Zarzian
tradition" to some part of the "Proto-Neolithic" where
there was the use of pressure technology.
For the time being, we do not háve absolute dates
from our sites and we must také a relative dating
approach. However, the TB75 and TB130 materials can
be compared with the "Proto-Neolithic" along the
Zagros flanks of North Iraq, whose stone industries háve
been well documented with clear stratigraphy. They are
represented by those of the open-air sites at Zawi Chemi
Shanidar, Qermez Dere, Nemrik 9, M'lefaat, Karim
Shahir and Maghzaliyah. Their artefact inventories are
outlined below.
Zawi Chem Shanidar is located in the Shanidar
valley, on the left bank of the Greater Zab River, north-
east of Mosul (Solecki 1981). Layer B of this site, dated
to 8,920 ±300 B.C. by radiocarbon determinations, was
associated with a circular stone architectural fearure
measuring 2 m. in diameter, and the lithic artefacts
consist of backed blades, denticulated pieces, notched
pieces, truncated pieces, borers, side-serapers, pieces
esquillées and geometrie microliths of the lunate
category. Neither sheen-bearing pieces nor micro-burins
were reported. The published photographs of the cores
demonstrate that the blades and microblades were not
detached by pressure.
Qermez Dere is located in the north- west outskirts of
the town of Telí Afar west of Mosul (Watkins et al.
1991). Seven settlement phases confírmed at this site
were grouped into three stages: the oldest stage (Phases
7 and 6) must be placed in the transition from the Epi-
Palaeolithic to the Neolithic periods, the middle stage
(Phases 5 and 4) without any Epi-Palaeolithic features
and belongs to the early Pre-Pottery Neolithic period,
and the youngest stage (Phases 3 and 2) also belongs to
the early Pre-Pottery Neolithic period. Except for some
geometrie microliths and micro-burins from the oldest
stage and the "Nemrik point" from Phase 4 onward, the
lithic artefacts bear the samé features through all the
phases, mainly consisting of notched pieces, denticulat-
ed pieces and Khiam points. The published illustrations
of the cores (Watkins et al. 1991; Fig. 8: 4) demonstrate
that pressure flaking was employed to detach blades and
microblades in the youngest stage at the latest.
Nemrik 9 is located in the southern part of the Dohuk
governorate, north-north-west of Mosul. Three
settlement phases confírmed at this site were placed in
the Pre-Pottery Neolithic period. Radiocarbon dates
pláce the oldest phase to the ninth millennium B.C., the
middle phase to the eighth millennium, and the youngest
phase to the end of the eighth millennium through the
first half of the seventh Millennium B.C. (Kozlowski
and Kempisty 1990: 349-50). The lithic artefacts show
essentially the samé features throughout the settlement
phases, that is, retouched blades, end-serapers, tanged
points called "Nemrik" points and microliths such as
backed microblades. Neither geometrie microliths nor
micro-burins are reported (Kozlowski and Szymczak
1990). It is stated that the blades were detached with the
aid of punches in the oldest phase and that they were
detached by pressure from the middle phase onward
(Kozlowski and Kempisty 1990: 350)
M'lefaat is located north-east of the Tigris-Greater
Zab junction on the west bank of the river Khazir, east
of the plain of Mosul (Dittemore 1983). Lithic artefacts
such as notched pieces, microblades with use-nibbled
edges, serapers and perforators were unearthed from
three stone made floors, 4 x 3 m. in area. These floors
were dated to between 8,900 and 8,600 B.C. (Howe
1983: 130-3 1). Neither geometrie microliths nor micro-
burins are reported. Parallel ridges and edges, and
consistent width of the microblades, demonstrate that
pressure flaking was employed to detach the
microblades at this site (Dittemore 1983: 673-74).
Karim Shahir is located east of Chemchemal in the
province of Kirkuk (Howe 1983). An enormous amount
of lithic artefacts were unearthed from a single
oceupation floor composed of stone pavement and pits
in the area of 500-600 m. 2 . This floor was dated to
between 8,900 and 8,600 B.C. (Howe 1983: 130-31).
20
JOURNAL OF PERSIAN STUDIES
The lithic artefacts consist of backed microblades, drills,
end-scrapers, side-scrapers, obliquely truncated pieces
and many notched pieces. Because geometrie microliths
of true category were not included, the micro-burins
fřom this site are considered as the by-products from the
manufacture of obliquely truncated pieces, unrelated to
the micro-burin technique. Most of the blade or
microblade cores are conical in shape, and their
published illustrations (Howe 1983: fígs 20, 21)
demonstrate that elaboráte pressure flaking was
employed to detach microblades.
Maghzaliyah is located some 12 km. west of Telí
Afar (Bader 1993). Strata 2 to 4 yielded several
rectangular houses. The lithic artefacts are made on
obsidian, except in the uppermost Stratům 4 with flint
artefacts slightly inereasing. These lithic artefacts bear
almost the samé features throughout the strata,
consisting of serapers, end-scrapers, borers, tanged
points, large flakes ušed unhafted as siekle blades, and
segmented blades to be inserted into sickles. Geometrie
microliths are very rare. Blade cores are conical in shape
or "bullet-shaped". Very regular flake scars on the cores
demonstrate that the blades were elaborately detached
by pressure flaking.
Apart from the "Proto-Neolithic" of North Iraq
briefly reviewed above, important materiál, comparable
to TB75 and TB130 materiál, was recently reported
from the Zagros Mountains in the province of Luristan,
West Irán. Roustaei and others carried out Palaeolithic
surveys there, and they published a report which
mentioned the Vare Zard site complex, a series of rock-
shelters with a scatter of Epi-Palaeolithic and "Proto-
Neolithic" artefacts extending for more than 200 m.
along the site complex on a cliff slope (Roustaei et al.
2004). Although these artefacts were collected during
surveys and not acquired through excavation, they
seem to show a defmite stone industry. Illustration of
the selected artefacts of this site complex interestingly
shows a strong similarity with our materials, both being
characterised by notched blades/microblades, end-
scrapers, borers and side-scrapers. Blades and
microblades were elaborately detached with parallel
ridges and edges, and microblades at least were
detached by pressure flaking. Some of the conical
microblade cores with regular flake scares can be
considered "bullet" cores (Roustaei et al. 2004: 705,
eg. ii).
In Fars province, especially the Marv Dasht plain,
many Palaeolithic cave sites háve been discovered.
Some of them, such as Eshkaft-e Gavi, were excavated
(Rosenberg 1985), and the Upper Palaeolithic lithic
assemblage was roughly detected. Epi-Palaeolithic
caves were extensively surveyed in the Marv Dasht
plain, and at least 1 1 cave sites, including a large cave
site named "KMC", produced typical latě Zarzian lithic
assemblages (Rosenberg 2003).
Many Portery Neolithic tappehs háve now been
excavated, in Fars province especially again in the Marv
Dasht plain. The excavations at the sites, such as Tal-i
Mushki and Jari B, revealed the earliest phases of the
TABLE 3. A Hypothetical Placement of the TB Sites in the "Proto-Neolithic" of the Zagros Flanks on the Basis of
Presence of Geometrie Microliths and Pressure Flaking of Microblades (modified from Ohnuma 1997: fig. 5).
WEST
EAST
Micro-burin technique -> < Epi-Palaeolithic Period >
Zawi Chemi Shanidar
Qermez Dere Phase 7
Qermez Dere Phase 6
Geometrie microliths
Qermez Dere Phase 5
Qermez Dere Phase 4 Nemrik 9 Oldest Phase <-
Qermez Dere Phase 3 Nemrik 9 Middle Phase
Qermez Dere Phase 2 Nemrik 9 Youngest Phase
Maghzaliyah
TB75 and TB130
• • • • Pressure flaking of microblades
M'lefaat Karim Shahir
Pottery
< Pottery Neolithic Period >
PROTO-NEOLITHIC CAVES IN THE BOLAGHI VALLEY
21
Portery Neolithic period in the samé region (Fukai et al
1973; Masuda 1986). Re-excavations on these early
Portery Neolithic sites háve recently focused on chrono-
logical clarifications (ex. Alizadeh et al 2004; Alizadeh
2006). Some newly investigated Portery Neolithic sites,
such as Toll-e Bashi and Kushk-e Hezar (Abdi et al
2003, Alden et al 2004) háve also shed new light on this
phase. The latě Zarzian lithic assemblage collected at
"KMC" consists of geometrie microliths, such as
lunates, curved backed bladelets, scalenes and quadrilat-
eral pieces, end-serapers and thumbnail-scrapers with
many blades and blade cores (Rosenberg 2003). The
lithic assemblage discovered from Tal-i Mushki, the
earliest Portery Neolithic site in this region, consists of
blades, denticulations, notched blades, lunates,
trapezoids, serapers, borers and others. Many siekle
elements were also included in this assemblage (Fukai et
al 1973).
VI. CONCLUSION
Taking all of the forgoing matters into consideration, we
tentatively pláce the lithic artefacts from Sites TB75 and
TB130 to some point in the "Proto-Neolithic"
framework of the Zagros mountains: particularly close
to the Qermez Dere Phases 5 and 4, Nemrik 9 Oldest
Phases, and perhaps M'lefaat and Karim Shahir, on the
basis of the presence of geometrie microliths and
pressure flaking of microblades (Table 3). This is
primarily due to the generál characteristics of the lithic
artefacts, i.e. the existence of the pressure flaking
technique, the absence of the micro-burin technique,
and the absence of siekle elements. This is an indication
of the existence of a transitional phase to the Neolithic
societies in the Eastern Zagros.
Some attributes of the lithic assemblages of TB75
and TB130 are common to those of the latě Zarzian, and
some other attributes are common to those from
Mushki. Despite this, the differences are much greater
than the similarities between these assemblages,
suggesting that the TB75 and TB130 assemblages are
coherent in and of themselves. As mentioned at the
beginning of this report, the transition between the end
of the Epi-Palaeolithic and the beginning of the Portery
Neolithic remains as a strange hiatus in the eastern
Zagros, and no other lithic assemblages identical to
those of TB75 and TB130 háve yet been found in the
eastern Zagros. We must progress with this project to the
next step in order to determine how the Eastern Zagros
might háve played a role in this important transition just
as the western Zagros region did.
Acknowledgements
We are grateful to Dr Masoud Azarnoush, the former
director of the Iranian Center for Archaeological
Research, for his kind permission and considerations to
our project. We also offer appreciation to Dr Hasan
Fazeli, the present director of ICAR. Mr Karim
Alizadeh, Ms Mojgan Seyedin and other staff of the
ICAR are always so helpful in allowing us to execute
our project. The local authority at Pasargadae, especially
Dr Mohamad Talebian, Director of the Parsa-
Pasargadae Research Foundation, offered us every
convenience possible. Financial assistance for the
research expenses of our project was provided by ICAR.
We are also grateful to the Japan Society for the
Promotion of Science which supported the travel
expenses of the Japanese members.
Bibliography
. Abdi, K., Pollock, S. and Bernbeck R. 2003. "Fars archaeology
project 2003: excavations at Toll-e Bashi", Irán 41 : 339-^4.
. Alden, J.R., Abdi, K., Azadi, A., Biglari, F. and Heydari, S. 2004
"Kushk-e Hezar: a Mushki/Jari period site in the Kur river
basin, Fars, Irán", Irán 42: 25-45.
Alizadeh, A. 2006. The Origins of State Organizations in
Prrehistoric Highland Fars, Southern Irán: Excavations at
Tall-e Bakun. University of Chicago Oriental Institute
Publications, Chicago
— , Zeidi, M., Askari, A., Niakan, L. and Atabaki, A. 2004
"Iranian prehistorie project: excavations at Tall-e Bakun A
and B, Jari A and B, and Mushki: reconstruction of the
prehistorie environment in Marvdasht", The Oriental
Institute 2003-2004 Annual Report 94-106. Chicago.
Ata'i , M. 2003. Preliminary report of the archaeological
survey in the Bolaghi Valley (in Persian).
Bader, N.O. 1993. "Telí Maghzaliyah: An Early Neolithic
Site in Northern Iraq". In N. Yoffee and J.J. Clark (eds),
Early Stages in the Evolution of Mesopotamian
Civilization. The University of Arizona Press, Tucson and
London: 7-40.
Dittemore, M. 1983. "The Soundings at M'lefaať\ In L.S.
Braidwood, RJ. Braidwood, B. Howe, CA. Reed and PJ.
22
JOURNAL OF PERSIAN STUDIES
Watson (eds), Prehistorie Archeology along the Zagros
Flanks. University of Chicago Oriental Institute
Publications, vol. 5: 671-92. Chicago.
Fukai, S., Horiuchi, K. and Matsutani, T. 1973. Marv-Dasht III
The Excavation at Tall i-Mushki 1965. Tokyo.
Howe, B. 1983. "Karim Shahir" In L.S. Braidwood, R.J.
Braidwood, B. Howe, CA. Reed and RJ. Watson (eds),
Prehistorie Archeology along the Zagros Flanks. University
of Chicago Oriental Institute Publications, vol. 5: 23-154,
Chicago
ICAR (Iranian Center for Archaeological Research). 2006.
Abstracts, Symposium on the Archaeological Rescue
Excavations in the Bolaghi Valley. ICHTO and ICAR,
Shiraz.
Kozlowski, S.K. and Kempisty, A. 1990. "Architecture of the
Pre-Pottery Neolithic Settlement in Nemrik, Iraq", World
Archaeology 21-3: 348-62.
Kozlowski, S.K. and Szymczak, K. 1990. "Flint Industry". In
S.K. Kozlowski, Nemrik 9: Pre-pottery Neolithic Site in
Iraq (General Report: Seasons 1985-1986). Warsaw
University Press, Warsaw: 59-103.
Maeda, A. 1986. "A study on the painted pottery from Tápe
Djari B", Bulletin of the Ancient Orient Museum 8:
55-86.
Ohnuma, K. 1997. "Chronology of the 'Proto-Neolithic' of Iraq
and Syria: A Hypothetical View", Al-Rafidayn 18: 45-58.
Olszewski, D.I. 1993. "The Zarzian Occupation at Warwasi
Rockshelter, Irán". In D.I. Olszewski and H.L. Dibble (eds),
The Paleolithic Prehistory ofthe Zagros-Taurus. University
Museum, University of Pennsylvania, Monograph 83:
207-36.
Rosenberg, M. 1985. "Report on the 1978 sondage at Eshkaft-e
Gavi"iran 23: 51-61.
— 2003. The Epipaleolithic in the Marvdasht. In Miller, N.F.
and Abdi, K. (eds.) Yeki bud, yeki nabud-Essays on the
Archaeology of Irán in Honor of William M. Sumner:
98-108, Cotsen Institute of Archaeology, University of
California, Los Angeles.
Roustaei, K, Vahdati Nasab, H., Biglari, E, Heydari, S., Clark,
G. A. and Lindly, J.M. 2004. "Recent Paleolithic Surveys in
Luristan", Current Anthropology 45-5: 692-707.
Solecki, R.L. 1981. An Early Village Site at Zawi Chemi
Shanidar Bibliotheca Mesopotamica, vol. 13. Undena
Publications, Malibu.
Watkins, T., Betts, A., Dobney, K, Nesbitt, M., Gale, R. and
Molleson, T. 1991. Qermez Dere, Tel Afar: Interim Report
No. 2, 1989. Project Paper No. 13. Department of
Archaeology, University of Edinburgh.
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Open Library
American Libraries
TV News
Understanding 9/11